Ferroelectric domain engineering and characterization for photonic applications

(1)

Ferroelectric domain engineering and characterization for photonic applications

Simonetta Grilli

Doctoral Thesis Department of Physics Royal Institute of Technology

Stockholm, Sweden 2006

(2)

Royal Institute of Technology Laser Physics and Quantum Optics Albanova

Roslagstullsbacken 21

SE-10691 Stockholm, Sweden

TRITA-FYS 2006:40 ISSN 0280-316X

ISRN KTH/FYS/--06:40--SE ISBN 91-7178-382-2

Cover picture: Optical diffraction pattern from a LiNbO3 crystal engineered with an hexagonal periodic surface structure.

Ferroelectric domain engineering and characterization for photonic applications

(3)

Grilli, Simonetta

Ferroelectric domain engineering and characterization for photonic applications Laser Physics and Quantum Optics, Department of Physics, Royal Institute of Technology,

SE – 10691 Stockholm, Sweden.

TRITA-FYS 2006:40 ISSN 0280-316X

ISRN KTH/FYS/--06:40--SE ISBN 91-7178-382-2

Abstract

Lithium niobate (LiNbO3) and KTiOPO4 (KTP) are ferroelectric crystals of considerable interest in different fields of optics and optoelectronics. Due to its large values of the nonlinear optical, electro-optic (EO), piezoelectric and acousto-optical coefficients, LiNbO3 is widely used for laser frequency conversion using the quasi- phase matching (QPM) approach where the sign of nonlinearity has been periodically modulated by electric field poling (EFP). In the microwave and telecommunication field LiNbO3 is used for surface acoustic devices and integrated optical modulators.

KTP and its isomorphs, on the other hand, exhibit slightly lower nonlinear coefficients but have much higher photorefractive damage thresholds, so that it is mainly used in the fabrication of QPM devices for both UV, IR and visible light generation and in high power applications.

This thesis focus on different key issues: (1) accurate characterization of specific optical properties of LiNbO3, which are of interest in nonlinear and EO applications;

(2) in-situ visualization and characterization of domain reversal by EFP in LiNbO3 and KTP crystals for a through understanding of the ferroelectric domain switching; (3) fabrication of periodic surface structures at sub-micron scale in LiNbO for photonic applications. An interferometric method is used for accurate measurement of ordinary and extraordinary refractive indices in uniaxial crystals, which is of great interest in the proper design of QPM crystals. A digital holography (DH) based method is presented here for 2D characterization of the EO properties of LiNbO , which is considerably interesting in the applications where the proper design of the EO device requires a spatially resolved information about the EO behaviour and the existing pointwise techniques are not sufficient. A DH method for novel in-situ monitoring of domain reversal by EFP in both LiNbO and KTP, is also presented here. The technqiue could be used as a tool for high fidelity periodic domain engineering but also provides information about domain kinetics, internal field and crystals defects.

3

Finally this thesis presents novel results concerning nanoscale periodic surface structuring of congruent LiNbO3. Holographic lithography (HL) is used for sub-micron period resist patterning and electric overpoling for surface domain reversal. Surface structures are obtained by selective etching. Moiré effect is also used in the HL to fabricate complicated structures with multiple periods. The depth compatibility with waveguide implementation allows foreseeing possible applications of these structures for Bragg gratings or innovative photonic crystal devices, exploiting the additional nonlinear and EO properties typical of LiNbO3.

Keywords: LiNbO , KTiOPO , interferometry, digital holography, electric field poling, electro-optic materials, holographic lithography, ferroelectric domains, nanostructures, microstructures.

3 4

(4)

(5)

To Luigi.

It pleases me as much to doubt as to know

Dante Alighieri, La Divina Commedia

(6)

(7)

Preface

The work presented in this thesis is the result of the doctoral studies started about four years ago in the Laser Physics and Quantum Optics Group, Department of Physics, at the Royal Institute of Technology (KTH). The experimental work has been performed partly in KTH and partly at the National Institute of Applied Optics (INOA) of the National Council of Research (CNR) in Naples, Italy, where I was holder of a Research Grant first and recently of a Researcher position.

The research work performed at the INOA laboratories have been funded by the Italian Ministero dell’Istruzione dell’Università e della Ricerca (MIUR) within the two Projects: Microdispositivi fotonici in niobato di litio (FIRB n. RBNE01KZ94); Circuiti fotonici integrati per le telecomunicazioni ottiche e la sensoristica (n. 77 DD n.

1105/2002).

This thesis consists of an introductory part providing a background to the work performed and the reprints of the publications.

(8)

(9)

List of publications

Publications included in the thesis

I. S. Grilli, P. Ferraro, S. De Nicola, A. Finizio, G. Pierattini, R. Meucci, “Whole optical wavefields reconstruction by digital holography,” Optics Express 9, 294 (2001). http://www.opticsinfobase.org/abstract.cfm?id=65208

II. S. De Nicola, P. Ferraro, A. Finizio, P. De Natale, S. Grilli, G. Pierattini, “A Mach-Zehnder interferometric system for measuring the refractive indices of uniaxial crystals,” Optics Communications 202, 9 (2002).

III. M. de Angelis, S. De Nicola, A. Finizio, G. Pierattini, P. Ferraro, S. Grilli, M.

Paturzo, L. Sansone, D. Alfieri, P. De Natale, “Two-dimensional mapping of electro-optic phase retardation in lithium niobate crystals by digital holography,”

Optics Letters 30, 1671 (2005).

IV. S. Grilli, P. Ferraro, M. Paturzo, D. Alfieri, P. De Natale, M. de Angelis, S. De Nicola, A. Finizio, G. Pierattini, “In-situ visualization, monitoring and analysis of electric field domain reversal process in ferroelectric crystals by digital holography,” Optics Express 12, 1832 (2004).

http://www.opticsinfobase.org/abstract.cfm?id=79763

V. S. Grilli, P. Ferraro, D. Alfieri, P. De Natale, C. Canalias, V. Pasiskevicius, F.

Laurell, “Characterization of fast dynamic evolution of ferroelectric domains in KTiOPO4 by digital holography,” submitted to Optics Express.

VI. C. Canalias, V. Pasiskevicius, F. Laurell, S. Grilli, P. Ferraro, P. De Natale, “In- situ visualization of domain kinetics in flux grown KTiOPO4 by digital holography,” submitted to Applied Physics Letters.

VII. M. de Angelis, S. De Nicola, A. Finizio, G. Pierattini, P. Ferraro, S. Grilli, M.

Paturzo, “Evaluation of the internal field in lithium niobate ferroelectric domains by an interferometric method,” Applied Physics Letters 85, 2785 (2004).

VIII. M. Paturzo, D. Alfieri, S. Grilli, P. Ferraro, P. De Natale, M. de Angelis, S. De Nicola, A. Finizio, G. Pierattini, “Investigation of electric internal field in congruent LiNbO by electro-optic effect,” Applied Physics Letters 85, 5652

(2004). ³

IX. M. Paturzo, P. Ferraro, S. Grilli, D. Alfieri, P. De Natale, M. de Angelis, A.

Finizio, S. De Nicola, G. Pierattini, F. Caccavale, D. Callejo, A. Morbiato, “On the origin of internal field in lithium niobate crystals directly observed by digital holography,” Optics Express 13, 5416 (2005).

X. S. De Nicola, P. Ferraro, A. Finizio, S. Grilli, G. Coppola, M. Iodice, P. De Natale, M. Chiarini, “Surface topography of microstructures in lithium niobate by digital holographic microscopy,” Measurement Science and Technology 15, 961 (2004).

XI. S.Grilli, P. Ferraro, S. De Nicola, A. Finizio, G. Pierattini, P. De Natale, M.

Chiarini, “Investigation on reversed domain structures in lithium niobate crystals patterned by interference lithography,” Optics Express 11, 392 (2003).

XII. S. Grilli, P. Ferraro, P. De Natale, B. Tiribilli, M. Vassalli, “Surface nanoscale periodic structures in congruent lithium niobate by domain reversal patterning and differential etching,” Applied Physics Letters 87, 233106 (2005) [selected for The Virtual Journal of Nanoscale Science & technology].

(10)

XIII. S. Grilli, P. Ferraro, L. Sansone, M. Paturzo, S. De Nicola, G. Pierattini, P. De Natale, “Double-face and submicron two-dimensional domain patterning in congruent lithium niobate,” IEEE Photonics Technology Letters 18, 541 (2006).

XIV. S. Grilli, C. Canalias, F. Laurell, P. Ferraro, P. De Natale, “Control of lateral domain spreading in congruent lithium niobate by selective proton exchange,”

accepted by Applied Physics Letters.

XV. G. Coppola, P. Ferraro, M. Iodice, S. De Nicola, S. Grilli, D. Mazzotti, P. De Natale, “Visualization of optical deflection and switching operations by a domain-engineered-based LiNbO electro-optic device,” Optics Express 11, 1212 (2003). http://www.opticsinfobase.org/abstract.cfm?id=72392

3

Publications not included in the thesis

A1. S. De Nicola, P. Ferraro, A. Finizio, S. Grilli and G. Pierattini, “Experimental demonstration of the longitudinal image shift in digital holography,” Optical Engineering 42, 1625 (2003).

A2. E. Allaria, S. Brugioni, S. De Nicola, P. Ferraro, S. Grilli and R. Meucci, “Digital Holography at 10.6 µm,” Optics Communications 215, 257 (2002).

A3. P. Ferraro, S. De Nicola, A. Finizio, G. Coppola, S. Grilli, C. Magro and G.

Pierattini, “Compensation of the inherent wave front curvature in digital holographic coherent microscopy for quantitative phase contrast imaging,”

Applied Optics 42, 1938 (2003).

A4. E. Allaria, S. Brugioni, S. De Nicola, P. Ferraro, S. Grilli and R. Meucci,

“Interferometric analysis of reorientational nonlinear phenomena at 10.6µm in a nematic liquid crystal,” Applied Optics 42, 4827 (2003).

A5. S. De Nicola, P. Ferraro, G. Coppola, A. Finizio, G. Pierattini, S. Grilli, “Talbot self-image effect in digital holography and its application to spectrometry,”

Optics Letters 29, 1 (2004).

A6. G. Coppola, P. Ferraro, M. Iodice, S. De Nicola, A. Finizio, S. Grilli, “A digital holographic microscope for complete characterization of microelectromechanical systems,” Measurement Science and Technology 15, 529 (2004).

A7. M. de Angelis, S. De Nicola, P. Ferraro, A. Finizio, S. Grilli, G. Pierattini, “Profile measurement of a one-dimensional phase boundary sample using a single shot phase-step method,” Optics and Lasers in Engineering 43, 1305 (2005).

A8. P. Ferraro, S. Grilli, D. Alfieri, S. De Nicola, A. Finizio, G. Pierattini, B. Javidi, G.

Coppola, V. Striano, “Extended focused image in microscopy by digital holography,” Optics Express 13, 6738 (2005).

A9. P. Ferraro, D. Alfieri, S. Grilli, L. Sansone, S. De Nicola, A. Finizio, G. Pierattini,

“Recovering correct phase information in multiwavelengths digital holographic microscopy by compensation of chromatic aberrations,” Optics Letters 30, 2706 (2005).

A10. S. Grilli, P. Ferraro, D. Alfieri, M. Paturzo, L. Sansone, S. De Nicola, P. De Natale, “Interferometric technique for characterization of ferroelectric crystals properties and microengineering process,” Fringe ’05, The 5 International Workshop on Automatic Processing of Fringe Patterns, pg. 514 (Springer, W.

Osten Ed.)

th

A11. S. Grilli, P. Ferraro, L. Sansone, M. Paturzo, S. De Nicola, A. Finizio, G.

Pierattini, P. De Natale, “Engineering and characterization of ferroelectric microstructures for photonic crystal applications,” Proc. SPIE 6182, 204 (2006) [Plenary Paper in the Session Hot Topics in Photonics].

(11)

5.2 In-situ characterization of domain switching in ferroelectric crystals 49 5.2.1 In-situ domain reversal imaging in LiNbO3 by the RGI 51 5.2.2 In-situ domain reversal imaging in LiNbO3 by the MZ 54 5.2.3 In-situ domain reversal imaging in KTiOPO by the MZ 4 56 5.2.4 Comparison of domain switching dynamics 59

5.3 Characterization of internal field in LiNbO3 60

(12)

5.3.1 Evaluation of the static internal field 60 5.3.2 Evaluation of the elastic component of the internal field 62 5.3.3 Evidence of the internal field dependence on crystal defects 65 5.4 Topography characterization of relief structures in LiNbO3 69

5.4.1 Principle of the method 70

5.4.2 Experimental results 71

References 76 Chapter 6. Ferroelectric domain engineering for photonic applications 77

6.1 Introduction 77

6.2 Holographic lithography 77

6.3 Sub-micron domains obtained by electric field overpoling 78 6.3.1 Overpoling regime applied to 1D gratings 79 6.3.2 Overpoling regime applied to 2D gratings 81 6.4 Periodic nanoscale surface structuring by electric field overpoling 83 6.5 Complex surface structures by moiré holographic lithography 85 6.6 Double-face sub-micron surface structures 88 6.7 Possible applications for novel photonic crystal devices 90 6.8 Periodic domain engineering by proton exchange 91

References 94 Chapter 7. Electro-optic device based on domain engineered LiNbO3 95

7.1 Beam deflector 97

7.2 TIR-based switch device 101

References 103 Chapter 8. Description of the original work and author contribution 105

Chapter 9. Summary 111

List of acronyms 115

Acknowledgments 117 Papers I - XV

(13)

Chapter 1 Introduction and motivations

1.1 Background

Lithium niobate (LiNbO or LN) and potassium titanyl phosphate (KTiOPO or KTP) are ferroelectric nonlinear crystals widely used in the field of optics and photonics.

LiNbO was reported to be ferroelectric by Matthias and Remeika in 1949 and was grown from a lithium fluoride flux [1]. Later it was also grown from the melt by using the Czochralski pulling technique by Ballman [2] and Fedulov [3]. The interest in this material started to increase in the 1990’s when a large number of technical publications related to LiNbO began to appear in literature. The rapid growth of photonics for communications, data storage, display, biomedical and defence applications resulted in an increasing need for a versatile solid-state platform with a range of linear and nonlinear optical properties, much as silicon in semiconductor industry. The crystal growth for LiNbO has been developed to an extent that high quality, single domain wafers of this material are nowadays commercially available at relatively low cost with diameters up to 4 inch and with significant homogeneity.

Nowadays LiNbO is widely used in the optical, laser and communications industry thanks to its high electro-optic (EO), piezoelectric (PZ), acousto-optic and nonlinear optical coefficients, transparency in a wavelength range from near UV to far infrared, large photovoltaic effect and photocurrents in doped crystals, and finally thanks to the possibility of patterning ferroelectric domains into various shapes and sizes [4]. In the field of lasers a not negligible second order nonlinear optical effect constitutes a valuable property, giving the possibility of realizing the key devices for building coherent sources emitting radiation from the UV to the IR spectral range [4,5]. This is achieved traditionally by birefringent phase-matching but in the last years much higher efficiency has been obtained in the generation of coherent radiation by using the quasi-phase-matched (QPM) interaction [6]. The laser sources emitting in the visible and UV find a wide variety of applications including fluorescent spectroscopy in bio-medicine, devices for printing and image recording, laser display systems and high-density optical storage. Those sources emitting in the near-to mid IR find main applications in the field of molecular spectroscopy including for example the development of gas sensing equipments. The operation of such QPM devices is based on the possibility of manipulating the ferroelectric crystals in order to produce ferroelectric domain structures, the so called “periodically poled” crystals, such as periodically poled LN (PPLN), which rank nowadays among the most efficient photonic components for frequency conversion applications [6]. The relatively high acousto-optic effect of LiNbO , combined with the PZ effect, gives the possibility of producing surface acoustic waves (SAW) to control light beams, with applications in the field of microwave communications as surface acoustic wave delay lines and filters [7]. In the field of fiber optic communication systems a wide variety of EO devices, such as switches and multiplexers where an electric field serves to control the light propagation along the medium, have been developed in LiNbO . In fact, LiNbO is the EO material which best addresses the high demands in terms of switching time currently required in signal processing components. In addition the well established technology (titanium indiffused or annealed proton exchange [8]) for

3 4

3

3 3

(14)

waveguide fabrication in LiNbO allows optical integration of those electro-optically controlled components. ³

KTP was first synthesized in 1890 by L. Ouvard [9] but it wasn’t until the 1970’s that Zumsteg, Bierlein and Gier identified the nonlinear optical properties of this crystal [10], which appear to be extremely desirable for several solid state laser applications. In fact, the unique combination of high nonlinear coefficients, high damage threshold and low photorefractive sensitivity makes it well suited for laser system applications requiring high power, high efficiency and/or durability. KTP can be phase matched at 1.06 µm but QPM frequency conversion processes are much more attractive for realizing nonlinear components. These, combined with diode- pumped solid-state lasers, allow the fabrication of compact and efficient new laser sources in the UV to the near-to-mid infrared wavelength range.

Periodically poled KTP (PPKTP) crystals present several advantages compared to the PPLN ones. The excellent nonlinearities of PPKTP crystals make them suitable for UV, visible and IR generation in high power applications thanks to the higher photorefractive damage threshold. The coercive field of KTP at room temperature is about one-order of magnitude lower than that of congruent LiNbO3 so that thicker periodically poled structures can be easily fabricated giving access to higher optical powers. Moreover the anisotropic lattice structure of KTP limits the domain broadening [11], thus favouring the fabrication of dense domain gratings. Anyway KTP still cannot become as popular as LiNbO for different reasons. First of all KTP crystals cannot be grown by Czochralski method so that the KTP wafers are not as large as those of LiNbO . The KTP material exhibits ionic conductivity which makes difficult the monitoring of the poling current during electric field domain inversion, thus requiring alternative monitoring techniques [11,12]. Such ionic conductivity varies from wafer to wafer and within the same wafer, thus reducing the yield of PPKTP crystals [13].

3

1.2 Motivations

Several current and future applications of LiNbO3 and KTP materials include efficient first-order second harmonic generation of short wavelengths in the UV range by QPM interaction, EO modulators based on Bragg diffraction and photonic crystal (PC) structures. The QPM interaction in submicron periodically poled nonlinear crystals would provide new coherent sources in the UV range which find applications in optical data storage and display systems. Compared to the conventional EO modulators, the Bragg grating based modulators would provide lower drive voltages and faster response times. In particular, Bragg grating structures operating at the typical telecom wavelengths would be of great interest to the field of telecommunications for the fabrication of different optical components such as electrically controlled Bragg reflectors, beam steering devices and narrow band filters. When a continuous electric field is applied over such a structure, the opposite domains acquire positive and negative refractive index changes, depending on their orientations, and a multilayer-index stack is generated. Such a layered refractive index structure can work as a narrow band filter, with the bandwidth and reflectivity depending upon the magnitude of the applied field and the length of the structure.

PCs implemented in nonlinear materials would provide additional functionalities in respect to the ones which are being developed in semiconductor substrates, with the new possibility to manipulate light by light itself [14]. Moreover the recent development of novel manufacturing techniques able to generate periodic surface

(15)

Introduction and motivations

structures in lithium niobate is attracting interest for novel applications of this material in the emerging field of photonic bandgap devices [15]. In fact, innovative PC devices based on LiNbO would be possible by using the additional functionalities related to its strong EO, PZ, nonlinear effects. However all of these applications require submicron domain periods that fall outside the current limit of routine fabrication capability and, as is often the case, a widespread technological application of these materials is preceded by a complete understanding of many of the fundamental physical processes occurring in these materials.

3

The motivations of this thesis were born in this research framework. This thesis focuses on different specific areas of interest including the characterization of specific properties of LiNbO3 and KTP crystals and the implementation of new techniques for ferroelectric domain engineering and surface structuring of LiNbO substrates with submicron periods. The novel characterization work presented in this thesis include accurate measurement of the ordinary and extraordinary refractive indices of LiNbO , two-dimensional mapping of the EO behaviour in LiNbO crystals, in-situ characterization of specific properties related to the electric field domain reversal in LiNbO and KTP crystals, such as domain switching kinetics and internal field. The method for measuring the refractive indices in uniaxial crystals is based on the rotation-dependent changes of the phase shift occurring in crystal slabs due to the variation of the optical path length. The measurements are based on interferometric type procedures and allow to evaluate ordinary and extraordinary indices with great accuracy compared to other techniques previously presented in literature [16-18].

Therefore this method is of great interest to the field of nonlinear applications where proper design of the nonlinear crystals for any specific application requires an accurate knowledge of the refractive indices [19,20]. The two-dimensional mapping of the EO induced phase shift in LiNbO crystals is performed by a digital holography (DH) based method and is motivated by the many efforts performed to investigate the EO properties of LiNbO due to its large involvement in the fabrication of EO controlled optical modulators. In fact, many techniques, such as polarimetric Senarmont type [21] and other interferometric methods [22] have been used for this purpose. Even though relatively highly sensitive measurements are provided by these techniques, the pointwise character of such measurements limit their usefulness to a restricted range of applications. For example, the methods in [22] and therein are useful as characterization tools in case of the bulk EO modulators used for telecom applications, where the EO response is of fundamental importance and its spatially resolved characterization is not strictly required. In fact the measured coefficient comes from either integration along the probe’s beam direction or from an average in its millimetre sized transverse section. On the contrary, information about the two-dimensional (2D) distribution and degree of uniformity of the EO response is required in LiNbO based photonic applications.

3

3 3

3

The characterization of properties related to the electric field poling is based on a DH technique where the linear EO and PZ effects occurring along the z crystal axis is used in order to image in-situ the domain switching dynamics. This technique would be useful for high fidelity control of domain patterning of ferroelectric crystals but also as a novel non-invasive tool for investigation of crystal properties. For example results concerning the characterization of domain kinetics in KTP crystals and internal field effects occurring in congruent LiNbO3 are presented in this thesis.

Finally this thesis presents the recent novel results concerning the fabrication of sub- micron period reversed domain gratings in congruent LiNbO3 with both 1D and 2D geometries by using a relatively easy to accomplish method based on holographic

(16)

lithography patterning and overpoling. Moreover, these domain structures are transferred into the substrate by simple wet etching in appropriate acid solutions in order to fabricate novel sub-micron surface structures which could find interesting applications in the field of PC applications.

This thesis is organized as follows. Chapter 2 provides the basic information about the LiNbO3 and KTiOPO4 crystals which are useful for the appropriate understanding of the topics treated in the successive chapters. Chapter 3 intends to provide a general knowledge of holography and particularly digital holography which is largely applied in this thesis for different high resolution characterizations. Chapter 4 presents the results concerning the accurate measurement of the ordinary and extraordinary refractive indices of LiNbO3 and those related to the 2D mapping of its EO coefficient. Chapter 5 focuses on the application of DH for the in-situ characterization of ferroelectric properties related to the electric field poling in case of both LiNbO3 and KTiOPO4. Chapter 6 reports the results obtained recently in the fabrication of sub-micron period reversed domain patterns in LiNbO and moreover of surface structures with 1D and 2D geometries. Chapter 7 describes the operation of two simple EO devices engineered in LiNbO crystals and functioning as beam deflector and optical switch. Chapter 8 describes the papers included in this thesis and the corresponding contribution by the author. Finally Chapter 9 provides a brief summary of this thesis.

3

(17)

Introduction and motivations

References Chapter 1

[1] B.T. Matthias, J.P. Remeika, Phys. Rev. 76, 1886 (1949).

[2] A.A. Ballman, J. Am. Ceram. Soc. 48, 112 (1965).

[3] S.A. Fedulov Z.I. Shapiro, P.B. Ladyzhenskii, Kristallografiya 10, 268 (1965).

[4] A. M. Prokhorov, Y.S. Kuziminov, Physics and Chemistry of Crystalline Lithium Niobate, Hilger, Bristol (1990).

[5] U. Simon, F.K. Tittel, in: F.B. Dunning, R.G. Hulet (Eds.), Atomic, Molecular and Optical Physics: Electromagnetic Radiation, vol. 29C, p. 231, Academic Press, New York (1997).

[6] M.M. Fejer, G. A. Magel, D. H. Jundt, and R. L. Byer, IEEE J. Quantum Electron.

28, 2631 (1992).

[7] A.A. Oliver (Ed.), Acoustic surface waves, Topics in Applied Physics, vol.24, Sringer (1978).

[8] E.L. Wooten, K.M. Kissa, A. Yi-Yan, E.J. Murphy, D.A. Lafaw, P.F. Hallemeier, D.

Maack, D.F. Attanasio, D.J. Fritz, G.J. McBrien, D.E. Bossi, IEEE J. Sel. Top.

Quantum Electron. 6, 69 (2000).

[9] L. Ouvrad, M. Troost, Compt. Rend. 121, 117 (1890).

[10] J.D. Bierlein, H. Vanherzeele, J. Opt. Soc. Am. B 6, 622 (1989).

[11] J. Hellström, R. Clemens, V. Pasiskevicius, H. Karlsson, and F. Laurell, J. Appl.

Phys. 90, 1489 (2001).

[12] H. Karlsson, F. Laurell, L.K. Cheng, Appl. Phys. Lett. 74, 1519 (1999).

[13] H. Karlsson, F. Laurell, Appl. Phys. Lett. 71, 3474 (1997).

[14] J.D. Joannopoulos, P. R. Villeneuve, and S. Fan, Nature 386, 143 (1997).

[15] K. Sakoda, Optical properties of photonic crystals, Springer Edition (2005).

[16] D.F. Heller, O. Kafri, J. Krasinnski, Appl. Opt. 33, 3037 (1985).

[17] J.C. Bhattacharya, Appl. Opt. 40, 1658 (2001).

[18] S. De Nicola, G. Carbonara, A. Finizio, G. Pierattini, Appl. Phys. B 58, 133 (1994).

[19] M.M. Fejer, G.A. Magel, D.H. Jundt, R.L. Byer, IEEE L. Quantum Electron. 28, 2631 (1992).

[20] F.D. Bloss, An introduction to the Method of Optical Crystallography, Holt, New York (1961).

[21] P.C. Lemaire, and M. P. Georges, Opt. Mater. 4, 182 (1995).

[22] P. Ney, A. Maillard, M. D. Fontana, and K. Polgàr, J. Opt. Soc. Am. B 17, 1158 (2000).

(18)

(19)

Chapter 2 General properties of LiNbO

₃

and KTiOPO

₄

2.1 Introduction

Lithium niobate (LiNbO3 or LN) is a human-made dielectric material that does not exist in nature. It was first discovered to be ferroelectric in 1949 [1] and it is now one of the most widely used electro-optic (EO) materials. LiNbO is characterized by large pyroelectric, piezoelectric (PZ), nonlinear and EO coefficients and has useful acoustic and acousto-optic properties. This richness of large-magnitude physical effects has caused LN to become widely used in applications such as acoustic wave transducers, optical amplitude modulators, second-harmonic generators, beam- deflectors, dielectric waveguides, memory elements, holographic data processing devices, and others [2]. Potassium titanyl phosphate (KTiOPO or KTP) is a non- centro-symmetric crystalline material which was first synthesised in 1890 [3]. It has been widely used in various nonlinear-optical applications, in particular in second harmonic generation (SHG) and optical parametric oscillation (OPO) devices based on pumping with 1 µm wavelength radiation from Nd lasers. This is due to a highly attractive combination of material properties including large nonlinearity, high damage threshold, large birefringent phasematching acceptance in angle and temperature, wide transmission, and good thermal and mechanical stability [4]. KTP and some of its isomorphs are available from several vendors and are today frequently used in commercial laser devices.

3

4

2.2 Structure and crystal growth Lithium niobate

The structure of LN at room temperature belongs to the 3m group and consists of planar sheets of oxygen atoms in a distorted hexagonal close-packed configuration [5]. Octahedral interstices are formed, one third of which is occupied by niobium atoms, one third by lithium atoms while the rest is vacant. Figure 2.2-1 shows the schematic view of the Li and Nb atoms position respect to the oxygen planes. Above the Curie temperature Tc (around 1210 °C) the phase is paraelectric (no spontaneous polarization) while in the ferroelectric phase below Tc LN exhibits spontaneous polarization. In the paraelectric phase the Li atoms lie in an oxygen layer that is c/4 away from the the Nb atom while the Nb atoms are centred between oxygen layers.

Conversely, in the ferroelectric phase the elastic forces of the crystal become dominant and force the lithium and niobium ions into new positions. Figure 2.2-2 shows the conventional hexagonal unit cell for LN, where the c axis is defined as the axis about which the crystal exhibits three-fold rotation symmetry [5]. The three equivalent a axes (a ,a ,a ) of the conventional hexagonal unit cell are 120° apart, lie in a plane normal to the c axis and are chosen to be perpendicular to the mirror planes of symmetry [5].

1 2 3

The ferroelectric dipole axis is aligned along the c axis, resulting in a c+ and a c- face, both of which are normal to the c axis. The c+ face corresponds to the positive end of the dipole which becomes negative upon compression. This is due to the free charges which deposit on the c faces to compensate the spontaneous polarisation,

(20)

so that negative charges deposit on the c+ face and positive charges on the c- face.

Under compression the niobium and lithium ions move closer to their paraelectric positions, reducing the intensity of the dipole. This results in excess of compensating charge on the two faces, so that the positive end of the dipole exhibits a negative charge under compression.

(a) (b) c+

axis

Fig. 2.2-1: Schematic view of the LiNbO3 crystal structure in (a) ferroelectric phase (below Tc) and (b) paraelectric phase (above Tc). [5]

A Cartesian coordinate system (x,y,z) is used to describe the physical properties of LN. Conventionally, the z axis is chosen to be parallel to the c axis, while the x axis is chosen to be parallel to one of the equivalent a axes. The y axis is then chosen so that a right handed system is formed, resulting in the y axis lying in one of the mirror symmetry planes.

Fig. 2.2-2: Schematic view of the hexagonal unit cell of LiNbO3 showing the hexagonal axes (c, a1, a2, a ), where the large circles are Nb ions and small circles are the Li atoms. 3

LN was first grown in the early 1960s using the Czochralski technique [6]. The pure compound of the material to be grown is melted in a crucible in the growth station and a seed crystal suspended from a rotating seed rod is lowered into the furnace cavity. Growth of a crystal can be initiated by starting to slowly withdraw the seed rod so that new crystal material attaches to the solid seed. The composition of a given crystal depends on the proportions used in the starting materials. For example, the congruent composition contains 48.45 mol% Li O and 51.55 Nb O [7]. In particular, stoichiometric LN (SLN), in common with many oxides, melts incongruently.

Differences between stoichiometric and congruent composition LN are apparent at

2 2 5

(21)

General properties of LiNbO₃ and KTiOPO₄

the unit cell level [8]. In SLN with T =1470 K, the Li atom site is 100% occupied by Li and the Nb site is 100% occupied by Nb, as the O site is 100% occupied by O [8]. A SLN composition can be obtained by using a lithium rich melt or by adding K O during the growth process. Anyway, the growth of defect-free SLN crystals is quite difficult. Both mechanical twinning [9] and lithium excess micro-clusters are possible due to the composition variation in the melt-crystal interface during the growth of off- congruent LN crystals. However, there is not enough experimental evidence that gives a thorough understanding of such defects. Lithium excess micro-clusters are generated during the crystallization process. They are commonly present in a limited area close to the interface and come from the bottom of the crucible from which they are transported by the natural convection currents. The clusters force the formation of lithium aggregates in the crystal lattice, generating micro-regions with a different lattice structure.

c

2

Potassium titanyl phosphate

KTiOPO4 belongs to the family of compounds that have the formula unit MTiOXO4, where M can be K, Rb, Tl, NH4, or Cs and X can be P or As [4]. All these materials are orthorhombic and belong to the acentric point group mm2. In case of KTP the structure is characterized by chains of TiO6 octahedra linked at two corners and by PO4 tetrahedra which separate the chains, as shown in Fig. 2.2-3.

(a) (b)

Fig. 2.2-3: Schematic view of the KTP crystal structure under (a) a-c projection and (b) a- b projection. Shaded elements are the Ti octahedra, open elements are the P tetrahedra, and open circles are the K. The short Ti-O bonds are shown as bold lines. [4]

There are two chains per unit cell. Alternating long and short Ti-O bonds occur along these chains generating a net z directed polarization and the major contribution to KTP's large nonlinear optic and EO coefficients. The K ions are weakly bonded to both the Ti octahedra and P tetrahedral, so that channels exist along the z axis whereby K can diffuse through a vacancy mechanism, with a diffusion coefficient several orders of magnitude greater than in the xy plane. Large single crystals of KTP can be grown by both flux and hydrothermal (Ht-KTP) techniques.

The flux technique is essentially a high-temperature solution growth process in which the KTP crystallizes out of a molten self-flux composition when cooled.

Depending on the specific flux used, crystal growth can occur from approximately 700°C to approximately 1000°C. The flux is a solvent of the components of the crystal. A significant advantage of the flux process is the operation at atmospheric pressure without requiring sophisticated pressure equipment. However, to avoid growth striations and flux inclusions, uniform temperature and high levels of temperature control are required.

(22)

The hydrothermal process consists of sealing nutrient and seed crystals in a gold tube, inserting the tube into a high-pressure and high-temperature autoclave and growing crystals at constant pressure and temperature. This technique yields crystals of very good uniformity and optical quality. However, the process is more complicated and takes longer time, thereby increasing the price and limiting the availability of such crystals. Moreover, due to the equipment restrictions related to the high temperatures and pressures required, the size of the grown crystals is limited.

Fig. 2.2-4: Schematic view of the KTP crystal morphology. [4]

The KTP crystal morphology is similar for both flux and hydrothermal growth processes, and specifics depend on seed size and orientation. Figure 2.2-4 shows the schematic view of the KTP crystal morphology.

2.3 Ferroelectric properties

Ferroelectric crystals such as LiNbO and KTiOPO are characterised by displaying two or more states of different polarisation in the absence of external electrical fields [10]. They are part of a larger group of crystals, the pyroelectrics, which all have a spontaneous polarisation (P ) not necessarily reversibly invertible. Thus all ferroelectrics are pyroelectric, although the contrary is not true. Similarly pyroelectrics are all part of the PZ crystal group, which is characterised by generating an electric field when subject to a mechanical strain. Thus ferroelectrics, such as LN and KTP, are also pyroelectric and piezoelectric. Ferroelectricity exists only in non- centrosymmetric crystals. In general a ferroelectric crystal does not exhibit the same polarisation orientation throughout. Each region with uniform polarisation is called

“domain” and the distribution of domains forms the domain structure of the crystal.

The spontaneous polarisation of a ferroelectric material usually decreases when heated and, as described in section 2.2, at the Curie temperature the crystal phase becomes paraelectric.

3 4

s

The spontaneous polarization Ps of a ferroelectric crystal may be reversed by the application of a suitable external electric field by means of the so called “electric field poling” (EFP) process (see section 2.5). The minimum field required for EFP is called the “coercive field” Ec, which varies with material and temperature. The spontaneous polarisation Ps can be defined by the amount of charge Q required to reverse a single domain crystal of a certain area A:

(2.3-1)

∫

⁼

= I_poldt AP_S

Q 2

(23)

where I is the charge transfer current flowing in the external circuit during EFP. In case of KTP, the anisotropic structure of the crystal lattice results in a velocity much higher for domain walls in the x-y plane than for walls in planes parallel to the z-axis.

This is a favourable feature since domain broadening is limited when fabricating periodically poled structures along the z-axis. Conversely, the isotropic structure of LN causes velocity of domain walls in x-y plane equivalent to that along the z axis.

Table 2.3-1 shows the values of the basic parameters related to the ferroelectric properties of LN and KTP crystals.

KTiOPO4

(positive biaxial)

nx=1.76 n =1.77 y

nz=1.86

pol

Table 2.4-1: Refractive index values of congruent LN [15] and flux-grown KTP [17]

measured at room temperature for 632.8 nm.

LN is transparent to wavelengths from around 400 nm to about 5.5 µm [18] while the transparency range of KTP is from 350 nm to 4.5 µm. Figure 2.4-1 shows the absorption spectrum measured for congruent LN and flux-grown KTP.

Tc (°C) P (c/ms ²) Ec (kV/mm)

LiNbO3 1210 0.71 21

KTiOPO4 946 0.14 2

Table 2.3-1: Values of the basic parameters related to the ferroelectric properties of LN and KTP. [11-13]

Built-in internal fields have been reported in ferroelectric crystals such as LiTaO3 [14]:

the spontaneous polarization measured as function of the applied external field exhibits an hysteresis loop. The presence of the so called “internal field” (IF) E is indicated by the large asymmetry in the loop. The magnitude of the IF can be determined empirically as E =(E -E )/2 where E and E are the electric fields required for first forward and reverse poling, respectively.

int

int f1 r1 f1 r1

2.4 Optical properties

LN is classified as a negative uniaxial crystal, having two different refractive indices for ordinary and extraordinary polarized light, n and n respectively, with n <n . Conversely KTP is a positive biaxial crystal with three different refractive index values n , n and n . Table 2.4-1 shows the refractive index values of congruent LN and flux- grown KTP crystals measured at room temperature for 632.8 nm. The index of refraction is also temperature dependent due to the thermo-optic effect. In case of LN the temperature dependant Sellmeier equations allow more accurate calculation of the refractive index [15]. In case of KTP the temperature derivative can be approximated by a Laurent series [16].

o e e o

x y z

no=2.29 ne=2.20 LiNbO3

(negative uniaxial)

(24)

(a)

(b)

Fig. 2.4-1: Typical transmission spectrum of (a) congruent LN [19] and (b) flux-grown KTP [13], where the transmission spectra of other crystals from the KTP family are also shown.

LiNbO3 and KTiOPO4 are EO crystals. The linear EO effect, also known as Pockels effect, is a linear change in the refractive index ellipsoid induced by an applied electric field. This change is produced by the first power of the field, in contrast with the quadratic or Kerr EO effect. The change in the refractive index ellipsoid is expressed as

(2.4-1)

where Ek are the components of the electric field and r represents the EO tensor which nonzero independent elements are determined by the crystal symmetry of the material. For example, in case of LN, the extraordinary index change derived from (2.4-1), when the electric field is applied along the crystal symmetry axis (z-axis), is

ijk

∑

=



 



∆

k ijk k ij

E n² r

1

(2.4-2)

The structure of the EO tensor for LN and KTP is given below:

3 3 33

2

1n r E n_e =− _e

∆

(25)

r (2.4-3)

[LN] [KTP]













−

=

0 0

22 42

42 33 13 22

13 22

r r

r r r r

r r

r_ij













=

0 0 0

0 0

51 42

33 23 13

r r

r r r

ij

The non vanishing EO tensor elements and their values for LN and KTP are listed in Table 2.4-2.

LiNbO3

r13=10 pm/V r =6.7 pm/V 22

r33=32.2 pm/V r₄₂=32.6 pm/V

KTiOPO4

r13=9.5 pm/V r =15.7 pm/V 23

r =36.6 pm/V 33

r42=9.3 pm/V r₅₁=7.3 pm/V

Table 2.4-2: Electro-optic tensor elements values for LN [20,21] and KTP [4] at room temperature for static electric field and light of 632.8 nm.

LN and KTP are the most widely used nonlinear crystals. Considering that the material response to electric fields is in general not linear, the electric polarisation can be expanded in powers of the field in the form:

(2.4-4)

The first right term is the usual linear dependence. The second term corresponds to the second order or quadratic nonlinearity, and next is the third order dependence.

Only non-centrosymmetric crystals, as is the case of LN and KTP, can possess a non-vanishing dijk tensor [22]. Usually in these materials the second order processes are much more important in magnitude than those of higher orders for the moderate electric field commonly present in the materials. The second order nonlinear susceptibility tensor structure for LN and KTP is given below:

...

4

2 ⁽³⁾

0 + + +

=

∑ ∑ ∑

j k l

jk ijk j k jkl ijkl j

j ij

i E d E E E E E

P ε χ χ

[LN]

















−

=

0 0

0

0 0

0

2 0

0 0 0

33 31 31

15 22

22

22 15

d d d

d d

d

d d

d_ijk

(26)

[KTP] d (2.4-5)

The main non-vanishing d second order nonlinear tensor elements for LN [23,24]

and KTP [25], measured at 532 nm, are presented in Table 2.4-3.

















=

0 0 0

0 0

0 0 0

33 32 31

32 31

d d d

d d

ijk

LiNbO3

d15=d =-4.35 pm/V 31

d22=2.10 pm/V d33=-27.2 pm/V

KTiOPO 4

d31=1.4 pm/V d =2.65 pm/V 32

d33=10.7 pm/V

Table 2.4-3: Second-order nonlinear tensor elements for LN and KTP crystals. [23-25]

A number of nonlinear optical effects are observed as a result of the second order interaction. For accounting of this effects it is common to use the following nonlinear polarization expression:

(2.4-6) When the two field components are of the same wave of frequency ω, the nonlinear

polarization will have a frequency 2ω, giving rise to the generation of a second harmonic new wave (SHG). Two different optical fields of frequencies ω1 and ω2 can be added to generate the sum frequency wave with ω3=ω +ω . In parametric amplification two waves of frequencies ω ^andω ^withω ^>ω and I >> I give rise to a new wave of frequency ω =ω −ω , and the amplification of the ω wave. In optical parametric oscillation (OPO) a wave of frequency ω is decomposed in frequencies ω ^andω such that ω1=ω2+ω3, provided the resonance of the system for at least one of the created frequencies. In order to obtain a practical intensity of the resulting wave, a phase matching condition is required for the new wave to be growing efficiently along the material in a coherent way. Phase matching conditions can be obtained in anisotropic materials for those propagation directions and polarizations for which the speed of waves of different frequencies are equal. For example in case of LN the exact birefringent phase matching condition for SHG is [22]

2 1 , 2 : 1 3

3 ω ω ω ω ω

ω ijk j k

i d E E

P =

1 2

1 2 1 2 1 2

3 1 2 2

1

2 3

(2.4-7)

ω

ω o

e n

n² =

This is a very restrictive condition that can be fulfilled only for certain frequencies, propagation directions and sample temperatures. By fortune there is another possibility, more easy to fulfil, to obtain efficient wave generation by the so called quasi phase matching (QPM) disposition [26]. This can be obtained in materials where a property has a spatial periodic change of the proper period. In LN and KTP crystals this can be easily obtained with periodic domain structures, as described in Section 2.5. QPM by a periodic material structure is a technique to maintain the relative phase between the interacting waves in the material. In fact this method

(27)

corrects the relative phase mismatch at regular intervals by means of a structural periodicity of the nonlinear medium. In case of LN and KTP crystals, periodically alternated domain structures have been produced by different methods. In SHG a fundamental wave of frequency ω1 and wavelength λ produces a polarization wave of frequency 2ω1 through the second order nonlinear susceptibility. This polarization wave is forced by the fundamental wave, and then it travels with the same velocity ν1=c/n1. The polarization wave radiates at each point of the material a free second harmonic optical wave, which travels with a velocity ν2=c/n2. Because of the normal dispersion of the material, in general is n2 > n1, so that the fundamental and the second harmonic waves travel at different phase velocities. The sign of the energy transfer from one wave to the other is determined by the relative phase between them. The continuous increment of relative phase difference leads to an alternation in the direction of the flow of power between the waves. This produces an alternating growth and decay of the power of the second harmonic wave, which in general never gets a practical intensity. Along a propagation distance over which the relative phase changes in π the power is transferred to the second harmonic wave, and in the next equal distance the phase changes to 2π and the power is transferred to the fundamental wave. This distance, known as the coherence length L , is given by c

(2.4-8)

Of course, when the refractive indices are equal for both waves, for example using the birefringence properties of the material, this coherence length becomes infinite.

(

2 1

)

4n n L_c

= λ−

(a) (b)

Fig. 2.4-2: Intensity of second harmonic generation as a function of the interaction length, (a) for exact birefringent phase matching, and (b) for quasi-phase matching in a periodically reversed LN sample with period 2Lc.

In this situation of phase matching the electric field of the second harmonic wave grows linearly with the interaction length, and thus the intensity grows quadratically (see Fig.2.4-2).

Another possibility to obtain a final practical second harmonic wave is to produce a repeated inversion of the relative phase between the waves. One way to invert the phase of the generated wave is to change the sign of the nonlinear coefficient. This is done in LN and KTP by inversion of the direction of the ferroelectric domains. The highest efficiency of SHG in this situation is obtained when the inversion is done every coherence length, i.e. a change period of 2Lc. This situation is illustrated in

(28)

Fig.2.4-2(b). Other longer periods can lead to a lower growth of the harmonic power.

As it can be seen in Fig. 2.4-2, the second harmonic power grows more slowly using QPM than it does with birefringent phase matching. The greatest advantage of QPM lies in its ability to produce phase matching in situations which would otherwise be impossible, for example the use of nonlinear coefficients coupling waves of the same polarization. Non-linear coefficients of highest value can then be addressed. As the domain period can be changed, another advantage is the possibility to design a device operating at the desired wavelengths and temperature [26]. Additionally, the acceptance bandwidths can be significantly enhanced, leading to less critical phase matching. These structures can also be used to phase match other second order effects as parametric amplification or oscillation. Several methods have been developed to produce periodic domain structures in LN (PPLN) and KTP (PPKTP).

The most widely used nowadays consists in the application of high external electric fields at room temperature to reverse the spontaneous polarization of the crystal, the so called electric field poling (EFP), as described in the next Section. This technique has produced PPLN [27-29] and PPKTP [30] for QPM SHG.

2.5 Periodic electric field poling

Starting from a single-domain crystal the production of periodic domain structures is achieved by application of electric fields at room temperature [27]. The challenge of fabricating high quality QPM structures by EFP lies in achieving few micrometers wide domains in crystals of several millimetres in thickness. This put high demands on the poling process. Most last years papers on PPLN structures obtained by EFP use commercial samples of 0.5 mm thickness, and interaction lengths of few centimetres are commonly reported. Thicker samples cannot be used because of the dielectric breakdown appearing before domain inversion. Recently it has been reported the decrease of the ferroelectric coercive field by one order of magnitude between congruent and stoichiometric crystals. This makes possible the electric field fabrication of periodic structures at room temperature in few millimetres thick samples [31].

The periodic EFP (PEFP) consists of using lithographic techniques to produce a photoresist grating of the desired period to be used as a mask for applying the electric field with a liquid electrolyte electrode [27]. A positive voltage pulse slightly exceeding the coercive field of the material (around 21 kV/mm in LN and 2 kV/mm in KTP) is applied on the patterned crystal face by using a liquid electrolyte. The liquid electrode configuration has two electrolyte containing chambers which squeeze the sample between two O-ring gaskets, as shown schematically in Fig. 2.5-1.

Fig. 2.5-1: Schematic view of the sample cross section after lithographic patterning.

crystal

Figure 2.5-2 illustrates the external electrical circuit. A conventional Signal Generator (SG) drives an High Voltage Amplifier (HVA - 2000x) with a series current limiting resistor Rs in order to get a 12kV positive voltage. A diode rectifier D is connected to the output of the HVA to prevent flowing of backswitch current in the circuit.

(29)

Fig. 2.5-2: Schematic view of the external circuit for EFP. SG signal generator; HVA high voltage amplifier; D diode rectifier; R series resistor; V poling voltage; HVP high voltage probe; R monitoring resistor; OSC oscilloscope; I poling current. m ^s pol^pol

In case of LN the reversed domains typically grow beyond the width of the electrodes as result of the remaining fringing fields along the edges of the lithographic grating strips [27]. For example, in PPLN processed for infrared applications (periods >10 µm), the inverted domain width will typically be ∼3-4 µm wider than that of the electrode. To obtain the desired domain size, insulating strips wider than the electrodes must be fabricated. The strategy for optimal domain patterning is to stop the voltage pulse before poling progresses under the photoresist layer. This is usually accomplished by delivering the apriori known amount of charge Q=2PsA required for polarization reversal in the electrode regions [32]. An in-situ stopping criterion consisting in watching for a drop in the poling current I and a corresponding rise in the poling voltage V , both effects indicating that the sample has completely poled under the electrodes and that the domains are now laterally spreading under the insulating layer, is used [27,32]. In fact, the conductivity of LN at room temperature is low enough that the poling current can be monitored readily by measuring the voltage drop across the R resistor (usually 10 kΩ) while a conventional High Voltage Probe (HVP) is used to measure the poling voltage V across the sample. Both current and voltage waveforms are visualized on the oscilloscope OSC during the poling process.

pol pol

m

pol

In case of KTP crystals the high ionic conductivity is a key-factor that provides the main difficulty in achieving PEFP of materials from the KTP family. High ionic conductivity results in a large current flow through the crystal when the high voltage is applied. Such high ionic conductivity makes it difficult to control the domain reversal by monitoring the poling current through the crystal. An alternative method based on the transverse EO effect in the crystal has been developed, where the time-dependent changes in the polarisation-state of a He-Ne laser beam that propagates through the crystal during poling are monitored [33].

Reversed ferroelectric domain patterns are usually inspected unambiguously by a well established technique based on selective wet etching [34]. Different etching solutions are used for LN and KTP with the property to attack the z- polarity face while leaving z+ face untouched. In case of LiNbO3 the etching solution consists of a HF:HNO =1:2 mixture, while in case of KTP the reversed domains are usually selectively etched by a KOH:KNO =2:1 mixture. Faster etching rates are obtained by acid bath at high temperature. Alternative approaches include the non-invasive domain visualization by crossed polarizers [35] and more recently the observation of domain structure by the EO effect [36].

3

(30)

2.6 Piezoelectric properties

LiNbO and KTiOPO are piezoelectric (PZ) crystals. The direct PZ effect consists in the formation of an electrical charge in proportion of an externally applied force.

Conversely, according to the inverse PZ effect, an electric field E induces a deformation of the crystal, described by the strain S (shrinking or expanding). The mathematical relations are similar to those encountered in case of the EO effect (see section 2.4):

3 4

(for direct effect) (2.6-1) and

(for inverse effect) (2.6-2)

where σjk are the components of the stress tensor, P are the components of the induced polarisation vector, d are the components of the PZ strain tensor, S are the components of the induced strain tensor and E are the components of the applied electric field.

i

ijk jk

i jk ijk jk

i d

P ⁼

∑

σ

∑

=

i ijk i

jk d E

S

LiNbO 3

d15= 69.2 10^-12C/N d22= 20.8 10^-12 C/N d31= -0.85 10^-12 C/N d33= 6.0 10^-12 C/N KTiOPO4 d31= 1 pm/V

d33= 27 pm/V

Table 2.4-3: Piezoelectric strain tensor elements for LN [38] and KTP crystals [39].

PZ materials are mostly used for positioning applications in devices such as atomic force microscopes (AFMs) and translation stages, where they allow very precise positioning with a resolution of nanometers [37], surface acoustic wave devices, where they allow to transfer an electrical signal to a surface acoustic wave, and pressure sensors. The PZ strain tensor d of LN and KTP has the same symmetry as that of their nonlinear susceptibility tensor (see Eq.(2.4-5)) and their values are listed in Table 2.6-1.

ijk

(31)

References Chapter 2

[1] B.T. Matthias, J.P. Remeika, Phys. Rev. 76, 1886 (1949).

[2] A. Yariv, Quantum electronics, John Wiley& Sons (1989).

[3] L. Ouvrad, M. Troost, Compt. Rend. 121, 117 (1890).

[4] J.D. Bierlein, H. Vanherzeele, J. Opt. Soc. Am. B 6, 622 (1989).

[5] R.S. Weis, T.K. Gaylord, Appl. Phys. A 37, 191 (1985).

[6] K. Nassau, H.J. Levinstein, G.M. Loiacono, J. Phys. Chem. Solids 27, 983 (1966).

[7] H.M. O’Bryan, P.K. Gallagher, C.D. Brandle, J. Am. Ceram. Soc. 68, 493 (1985).

[8] S.C. Abrahams, P. Marsh, Acta Crystallogr. B 42, 61 (1986).

[9] M. Park, K.Kitamura, K. Terabe, Y. Furukawa, Y. Ji E. Suzuki, J. Crys. G. 180, 101 (1997).

[10] M.E. Lines, A.M. Glass, Clarendon Press Oxford (1977).

[11] K. Nassau, H.J. Levinstein, G.M. Loiacono, J. Phys. Chem. Solids 27, 989 (1966).

[12] I. Camlibel, J. Appl. Phys. 40, 1690 (1969).

[13] H. Karlsson, PhD thesis, Department of Laser Physics and Quantum Optics, The Royal Institute of Technology, Stockholm.

[14] V. Gopalan, M.C. Gupta, J. Appl. Phys. 80, 6099 (1996).

[15] D.S. Smith, H.D. Riccius, R.P. Edwin, Opt. Comm. 17, 332 (1976).

[16] W. Wiechmann, S. Kubota, T. Fukui, H. Masuda, Opt. Lett. 18, 1208 (1993).

[17] H.Y. Shen, Y.P. Zhou, W.X. Lin, Z.D. Zeng, R.R. Zeng, G.F. Yu, C.H. Huang, A.D. Jiang, S.Q. Jia, D.Z. Shen, IEEE J. 28, 48 (1992).

[18] G.D. Boyd, R.C. Miller, K. Nassau, W.L. Bond, A. Savage, Appl. Phys. Lett. 5, 234 (1964).

[19] K.K. Wong Ed., Properties of lithium niobate (2002).

[20] R. Nevado, C. Sada, F. Segato, F. Caccavale, A. Kling, J.C. Soares, E. Cantelar, F. Cussò, G. Lifante, Appl. Phys. B 73, 555 (2001).

[21] N. Hamelin, Y.T. Chow, J. Mod. Opt. 45, 2125 (1998).

[22] A. Yariv, P. Yeh, Optical waves in crystals, Wiley series in pure and applied optics (John Wiley and Sons, New York) (1984).

[23] R.C. Miller, W.A. Nordland, P.M. Bridenbaugh, J. Appl. Phys. 42, 4145 (1971).

[24] U. Ellenberger, R. Weber, J.E. Balmer, B. Zysset, D. Ellgehausen, G.D. Mizell, Appl. Opt. 31, 7563 (1992).

[25] B. Boulanger J.P. Feve, G. Marnier, B. Menaert, X. Cabirol, P Villeval, C. Bonnin, J. Opt. Soc. Am. B 11, 750 (1994).

[26] M.M. Fejer, G. A. Magel, D. H. Jundt, and R. L. Byer, IEEE J. Quantum Electron.

28, 2631 (1992).

Ferroelectric domain engineering and characterization for photonic applications